gaze-following and recognizing intentions from gaze Outline - - PowerPoint PPT Presentation

gaze-following and recognizing intentions from gaze

Outline

• infant gaze following studies and

intentionality

• gaze following and object processing

Do infants gaze-follow?

• Infants turn in the direction that an adult has

turned.

• What is the mechanism underlying this behavior?
• The infant notices the head motion and swings

her head to the correct half of space without processing adults gaze to an object

• The infant follow the gaze and has an

understanding about the relationship between the

eyes open/closed experiment

• 12, 14, 18 month old infants
• between subject design
• conditions: adult’s closed or open eye
• Infant’s first target look was categorized as

“correct look” (+1) if it aligned with adult’s target and as “incorrect look” (-1) if it didn’t.

main results

• 10. EYES WIDE SHUT

22

1

tant from the infant, at approximately a 75" angle off midline. There were four trials (two to the left and two to the right in a counterbalanced order), and each trial was 6.5 s in duration. Thus, there were no linguistic or emotional cues as to where to turn, and no sound-localization cues because the targets were silent. The infant's behavior was videotaped and subsequently scored by an observer who remained blind to whether the adult turned with open or closed eyes and the direction of the adult turn. For each trial, an infant's first target look was categorized as a "correct look," when it aligned with the adult's target (+ l), or an "incorrect look," when it aligned with the opposite target

(- 1). If infants looked at neither target, they received a score of 0 for "non-

looking." As is standard in gaze-following procedures, the looking score was a total of the correct looks, incorrect looks, and non-looks (e.g., Butler, Caron, &Brooks, 2000; Flom, DeAk, Phill, & Pick, 2004; Moore & Corkum, 1998). Thus, if an infant consistently looks at correct targets, she would have a posi- tive score (with a maximum of 4), but if she frequently looks at incorrect tar- gets, her score would be negative (with a minimum of -4). The main findings are shown in Figure 10-1. Infants at all ages looked significantly more often at the target when the adult turned with open than with closed eyes. We also scored other behaviors beyond the traditional look- ing measure. We scored infants' average duration of correct looks. This re- vealed that infants inspected the target longer when the adult turned to it with

• pen versus closed eyes. Also, more infants vocalized toward the correct tar-

get in the open-eyes than closed-eyes condition. Finally, significantly more

Open Eyes Closed Eyes All Ages 12mo. 14mo. 18mo.

FIGURE 10-1. Infants look at the correct target more often in the open-eyes than the closed-eyes condition. (From Brooks & Meltzoff, 2002. Reprinted with permission of the American Psychological Association.)

• ther findings
• infants inspected the target longer in open-

eyes condition.

• more infants vocalized toward the correct

target in the open-eyes than closed-eyes condition.

• significantly more infants pointed to the

targets in open-eyes that closed-eyes

• condition. (interpreted as evidence of

“proto-declarative” pointing)

Is closed-eyes condition disruptive?

• The adult's eyes were shut only slightly longer than the

blink of an eye (half a second) before the turning toward the object for a 6.5 s response period.

• no detectable difference in the emotional reactions as a

function of condition

• The duration measures the length of looking after the

infant has turned to the correct target.

• Infants show other target-directed acts (pointing at the

target and vocalizing toward it) when the adult can see the target.

When does this begin?

• The same procedure

was used for 9 to 11 month old infants.

• 10. EYES WIDE SHUT

225

A METAMORPHOSIS IN GAZE-FOLLOWING IN INFANCY Rationale

The previous study showed that 12-month-olds gaze follow. The question re- mains as to when does this begin? The Gaze-Following: Eyes OpenIClosed test provides a tool for looking at the ontogenesis of gaze-following before the child's first birthday. Brooks and Meltzoff (2005) recently completed a study of infants during the tadpole era, from 9 months to 11 months of age. We used the same proce- dure as previously described, but tested infants within a remarkably controlled age window. The infants were recruited to fall at three discrete ages: 9, 10, and 11 months old, with each infant + 1 week of the target age. This was the equivalent of a cross-sectional microgenetic study-we assessed infants at three moments over a 90-day growth period to see if we could capture a meta- morphosis in behavior.

Empirical Findings and Interpretation

As shown in Figure 10-3, 9-month-olds did not discriminate between the

• pen- versus closed-eyes conditions. They turned equally often in both cases.

However, there was a clear developmental shift 30 days later. For 10-month-

• lds, the looking scores in the open-eyes condition were significantly greater

than in the closed-eyes condition; and a similar significant effect was also evi- dent among l l-month-olds. We also analyzed whether infants vocalized while looking at the correct target, categorized as a "correct gaze + simultaneous vocalization." (Infants

FIGURE 10-3.

At 9 months of age infants turn indiscriminately to the target, whether or not the adult can see it. But at 10 and

11 months old, they selectively

follow the gaze of the adult in the

• pen-eyes condition. Note the

sharp decline in looking when the adult cannot see the target (eyes closed). (From Brooks & Melt- zoff, 2005. Reprinted with permis- sion of Blackwell Publishing.)

Open Eyes Closed Eyes

9 10 11

Age in Months

What could the results for 9-month-old mean?

• they are limited to tracking adult’s head

movements and run into object by chance. (Butterworth)

• They are conditioned to the head

movements as a signal for seeing an object

• n the periphery. (Moore)
• They are body-orientation followers.

(Melrzoff's "Like Me" hypothesis) body

biological vs physical

• ccluders
• 12, 14, 18 month old infants
• conditions : headband, blindfold
• 14 and 18 month old infants looked at the

adult’s target significantly more often in the headband than in the blindfold condition.

• The 12 month old didn’t distinguish

between conditions.

intervention experiment

• Infants randomly assigned to a baseline

condition, or two treatment groups : blindfolds, and the same cloth with an

• pening cut in the middle of it.
• The blindfold group experienced that the

blindfold blocks their view.

• Those infants now interpreted the blindfold
• correctly. (like-me hypothesis?)

Does gaze-following behavior at 10-11 months predict later language development?

• Infants who produced the correct gaze and

simultaneous vocalization act at 10-11 months had larger receptive vocabulary at 18 months.

• They also built significantly more complex

sentences and had larger productive vocabulary at 24 months.

summary

• Gaze following happens in 10-11th month
• f infants development.
• Infants come to understand nonbiological
• ccluders to vision sometime around or

soon after 1 year of age depending upon the nature of the occluder.

• Gaze-following behavior at 10-11 months

predicts later language development.

gaze following and

• bject processing
• From gaze of others we get information

about both ‘inside’ and ‘outside’. (e.g. intentions and mental state vs information about relevant events in the environment)

• Is it possible to integrate the two groups?
• object properties

motor properties

emotional properties

• Do we prefer objects that are looked at by
• ther people?
• Fig. 1. Illustration of an experimental trial. Here, the face looks left (Box 3), validly cueing the eventual

emotional properties

status properties

• Does an object looked at by others look

more familiar?

• Fig. 1. Schematic of stimuli. (A) Direct gaze; (B) eye movement (C) final gaze; (D) central attractor and (E) presentation of two objects.

status properties

beyond an attentional shift

• experiments with

symbolic cues, like arrows, do not guarantee a consequence on object

• processing. (e.g. no

modulation observed in affective response to

• bjects using arrow

cues)

3 4 5 6 3 4 5 6

Average Rating Average Rating Experiment 1 – Gaze Cue

Congruent Incongruent Congruent Incongruent

Experiment 2 – Arrow Cue

beyond an attentional shift

• Non-intentional gaze does not produce the

same effects :

• fixating on the target and not looking at a

distractor

• looking away

beyond an attentional shift

• The properties discussed seem to persist

even when the gaze of others is no longer visible.

Summary

• Studies discussed suggest that gaze of
• thers modifies properties of objects and

influences object processing:

• object gazed at become graspable,

attractive and familiar.

• These properties are likely to be a

product of intentionality of the gaze.

References

• Meltzoff, A. N., & Brooks, R. (2007). Eyes wide shut: The importance of eyes in infant gaze following and

understanding other minds.

• Becchio, C., Bertone, C., and Castiello, U. (2008). How the gaze of others influences object processing.
• Bayliss, A.P

. et al. (2006) Gaze cueing and affective judgments of objects: I like what you look at.

• Bayliss, A.P

. et al. (2007) Affective evaluations of objects are influenced by observed gaze direction and emotional expression.

• Hayes, A.E. et al. (2007) Self produced and observed actions influence emotion: the roles of action

fluency and eye gaze.

• Reid,

V.M. and Striano, T. (2005) Adult gaze influences infant attention and object processing: implications for cognitive neuroscience.

Questions?